Skip to content

D
DeepSpeech

PaddlePaddle / DeepSpeech
大约 2 年 前同步成功

通知 210
Star 8425
Fork 1598
D
DeepSpeech
“a337e5b4779fd5e8821080dc67fea42c5018c83c”上不存在“doc_cn/api/v2/fluid/param_attr.html”
未验证 提交 f3ec985a 编写于 作者: H Hui Zhang 提交者： GitHub
浏览文件
操作

Merge pull request #1439 from Jackwaterveg/tipc 

[TIPC]Add tipc_benchmark of conformer
上级 ebef10ef aefe9e93
隐藏空白更改
内联 并排
Showing with 976 addition and 4 deletion
paddlespeech/s2t/exps/u2/model.py
浏览文件 @ f3ec985a
...@@ -175,7 +175,7 @@ class U2Trainer(Trainer): ...@@ -175,7 +175,7 @@ class U2Trainer(Trainer):
observation['batch_cost'] = observation[ observation['batch_cost'] = observation[
'reader_cost'] + observation['step_cost'] 'reader_cost'] + observation['step_cost']
observation['samples'] = observation['batch_size'] observation['samples'] = observation['batch_size']
observation['ips,sent./sec'] = observation[ observation['ips,samples/s'] = observation[
'batch_size'] / observation['batch_cost'] 'batch_size'] / observation['batch_cost']
for k, v in observation.items(): for k, v in observation.items():
msg += f" {k.split(',')[0]}: " msg += f" {k.split(',')[0]}: "
......
paddlespeech/s2t/training/trainer.py
浏览文件 @ f3ec985a
...@@ -252,8 +252,7 @@ class Trainer(): ...@@ -252,8 +252,7 @@ class Trainer():
if self.args.benchmark_max_step and self.iteration > self.args.benchmark_max_step: if self.args.benchmark_max_step and self.iteration > self.args.benchmark_max_step:
logger.info( logger.info(
f"Reach benchmark-max-step: {self.args.benchmark_max_step}") f"Reach benchmark-max-step: {self.args.benchmark_max_step}")
sys.exit( sys.exit(0)
f"Reach benchmark-max-step: {self.args.benchmark_max_step}")
def do_train(self): def do_train(self):
"""The training process control by epoch.""" """The training process control by epoch."""
...@@ -282,7 +281,7 @@ class Trainer(): ...@@ -282,7 +281,7 @@ class Trainer():
observation['batch_cost'] = observation[ observation['batch_cost'] = observation[
'reader_cost'] + observation['step_cost'] 'reader_cost'] + observation['step_cost']
observation['samples'] = observation['batch_size'] observation['samples'] = observation['batch_size']
observation['ips[sent./sec]'] = observation[ observation['ips samples/s'] = observation[
'batch_size'] / observation['batch_cost'] 'batch_size'] / observation['batch_cost']
for k, v in observation.items(): for k, v in observation.items():
msg += f" {k}: " msg += f" {k}: "
......
tests/test_tipc/benchmark_train.sh 0 → 100644
浏览文件 @ f3ec985a
#!/bin/bash
source test_tipc/common_func.sh
# set env
python=python
export model_branch=`git symbolic-ref HEAD 2>/dev/null | cut -d"/" -f 3`
export model_commit=$(git log|head -n1|awk '{print $2}')
export str_tmp=$(echo `pip list|grep paddlepaddle-gpu|awk -F ' ' '{print $2}'`)
export frame_version=${str_tmp%%.post*}
export frame_commit=$(echo `${python} -c "import paddle;print(paddle.version.commit)"`)
# run benchmark sh
# Usage:
# bash run_benchmark_train.sh config.txt params
# or
# bash run_benchmark_train.sh config.txt
function func_parser_params(){
strs=$1
IFS="="
array=(${strs})
tmp=${array[1]}
echo ${tmp}
}
function func_sed_params(){
filename=$1
line=$2
param_value=$3
params=`sed -n "${line}p" $filename`
IFS=":"
array=(${params})
key=${array[0]}
value=${array[1]}
if [[ $value =~ 'benchmark_train' ]];then
IFS='='
_val=(${value})
param_value="${_val[0]}=${param_value}"
fi
new_params="${key}:${param_value}"
IFS=";"
cmd="sed -i '${line}s/.*/${new_params}/' '${filename}'"
eval $cmd
}
function set_gpu_id(){
string=$1
_str=${string:1:6}
IFS="C"
arr=(${_str})
M=${arr[0]}
P=${arr[1]}
gn=`expr $P - 1`
gpu_num=`expr $gn / $M`
seq=`seq -s "," 0 $gpu_num`
echo $seq
}
function get_repo_name(){
IFS=";"
cur_dir=$(pwd)
IFS="/"
arr=(${cur_dir})
echo ${arr[-1]}
}
FILENAME=$1
# copy FILENAME as new
new_filename="./test_tipc/benchmark_train.txt"
cmd=`yes|cp $FILENAME $new_filename`
FILENAME=$new_filename
# MODE must be one of ['benchmark_train']
MODE=$2
PARAMS=$3
# bash test_tipc/benchmark_train.sh test_tipc/configs/det_mv3_db_v2_0/train_benchmark.txt benchmark_train dynamic_bs8_null_DP_N1C1
IFS=$'\n'
# parser params from train_benchmark.txt
dataline=`cat $FILENAME`
# parser params
IFS=$'\n'
lines=(${dataline})
model_name=$(func_parser_value "${lines[1]}")
# 获取benchmark_params所在的行数
line_num=`grep -n "train_benchmark_params" $FILENAME | cut -d ":" -f 1`
# for train log parser
batch_size=$(func_parser_value "${lines[line_num]}")
line_num=`expr $line_num + 1`
fp_items=$(func_parser_value "${lines[line_num]}")
line_num=`expr $line_num + 1`
epoch=$(func_parser_value "${lines[line_num]}")
line_num=`expr $line_num + 1`
profile_option_key=$(func_parser_key "${lines[line_num]}")
profile_option_params=$(func_parser_value "${lines[line_num]}")
profile_option="${profile_option_key}:${profile_option_params}"
line_num=`expr $line_num + 1`
flags_value=$(func_parser_value "${lines[line_num]}")
# set flags
IFS=";"
flags_list=(${flags_value})
for _flag in ${flags_list[*]}; do
cmd="export ${_flag}"
eval $cmd
done
# set log_name
repo_name=$(get_repo_name )
SAVE_LOG=${BENCHMARK_LOG_DIR:-$(pwd)} # */benchmark_log
mkdir -p "${SAVE_LOG}/benchmark_log/"
status_log="${SAVE_LOG}/benchmark_log/results.log"
# The number of lines in which train params can be replaced.
line_python=3
line_gpuid=4
line_precision=6
line_epoch=7
line_batchsize=9
line_profile=13
line_eval_py=24
line_export_py=30
func_sed_params "$FILENAME" "${line_eval_py}" "null"
func_sed_params "$FILENAME" "${line_export_py}" "null"
func_sed_params "$FILENAME" "${line_python}" "$python"
# if params
if [ ! -n "$PARAMS" ] ;then
# PARAMS input is not a word.
IFS="|"
batch_size_list=(${batch_size})
fp_items_list=(${fp_items})
device_num_list=(N1C4)
run_mode="DP"
else
# parser params from input: modeltype_bs${bs_item}_${fp_item}_${run_mode}_${device_num}
IFS="_"
params_list=(${PARAMS})
model_type=${params_list[0]}
batch_size=${params_list[1]}
batch_size=`echo ${batch_size} | tr -cd "[0-9]" `
precision=${params_list[2]}
# run_process_type=${params_list[3]}
run_mode=${params_list[3]}
device_num=${params_list[4]}
IFS=";"
if [ ${precision} = "null" ];then
precision="fp32"
fi
fp_items_list=($precision)
batch_size_list=($batch_size)
device_num_list=($device_num)
fi
IFS="|"
for batch_size in ${batch_size_list[*]}; do
for precision in ${fp_items_list[*]}; do
for device_num in ${device_num_list[*]}; do
# sed batchsize and precision
func_sed_params "$FILENAME" "${line_precision}" "$precision"
func_sed_params "$FILENAME" "${line_batchsize}" "$MODE=$batch_size"
func_sed_params "$FILENAME" "${line_epoch}" "$MODE=$epoch"
gpu_id=$(set_gpu_id $device_num)
if [ ${#gpu_id} -le 1 ];then
run_process_type="SingleP"
log_path="$SAVE_LOG/profiling_log"
mkdir -p $log_path
log_name="${repo_name}_${model_name}_bs${batch_size}_${precision}_${run_process_type}_${run_mode}_${device_num}_profiling"
func_sed_params "$FILENAME" "${line_gpuid}" "0" # sed used gpu_id
# set profile_option params
tmp=`sed -i "${line_profile}s/.*/${profile_option}/" "${FILENAME}"`
# run test_train_inference_python.sh
cmd="bash test_tipc/test_train_inference_python.sh ${FILENAME} benchmark_train > ${log_path}/${log_name} 2>&1 "
echo $cmd
eval $cmd
eval "cat ${log_path}/${log_name}"
# without profile
log_path="$SAVE_LOG/train_log"
speed_log_path="$SAVE_LOG/index"
mkdir -p $log_path
mkdir -p $speed_log_path
log_name="${repo_name}_${model_name}_bs${batch_size}_${precision}_${run_process_type}_${run_mode}_${device_num}_log"
speed_log_name="${repo_name}_${model_name}_bs${batch_size}_${precision}_${run_process_type}_${run_mode}_${device_num}_speed"
func_sed_params "$FILENAME" "${line_profile}" "null" # sed profile_id as null
cmd="bash test_tipc/test_train_inference_python.sh ${FILENAME} benchmark_train > ${log_path}/${log_name} 2>&1 "
echo $cmd
job_bt=`date '+%Y%m%d%H%M%S'`
eval $cmd
job_et=`date '+%Y%m%d%H%M%S'`
export model_run_time=$((${job_et}-${job_bt}))
eval "cat ${log_path}/${log_name}"
# parser log
_model_name="${model_name}_bs${batch_size}_${precision}_${run_process_type}_${run_mode}"
cmd="${python} ${BENCHMARK_ROOT}/scripts/analysis.py --filename ${log_path}/${log_name} \
--speed_log_file '${speed_log_path}/${speed_log_name}' \
--model_name ${_model_name} \
--base_batch_size ${batch_size} \
--run_mode ${run_mode} \
--run_process_type ${run_process_type} \
--fp_item ${precision} \
--keyword ips: \
--skip_steps 2 \
--device_num ${device_num} \
--speed_unit samples/s \
--convergence_key loss: "
echo $cmd
eval $cmd
last_status=${PIPESTATUS[0]}
status_check $last_status "${cmd}" "${status_log}"
else
IFS=";"
unset_env=`unset CUDA_VISIBLE_DEVICES`
run_process_type="MultiP"
log_path="$SAVE_LOG/train_log"
speed_log_path="$SAVE_LOG/index"
mkdir -p $log_path
mkdir -p $speed_log_path
log_name="${repo_name}_${model_name}_bs${batch_size}_${precision}_${run_process_type}_${run_mode}_${device_num}_log"
speed_log_name="${repo_name}_${model_name}_bs${batch_size}_${precision}_${run_process_type}_${run_mode}_${device_num}_speed"
func_sed_params "$FILENAME" "${line_gpuid}" "$gpu_id" # sed used gpu_id
func_sed_params "$FILENAME" "${line_profile}" "null" # sed --profile_option as null
cmd="bash test_tipc/test_train_inference_python.sh ${FILENAME} benchmark_train > ${log_path}/${log_name} 2>&1 "
echo $cmd
job_bt=`date '+%Y%m%d%H%M%S'`
eval $cmd
job_et=`date '+%Y%m%d%H%M%S'`
export model_run_time=$((${job_et}-${job_bt}))
eval "cat ${log_path}/${log_name}"
# parser log
_model_name="${model_name}_bs${batch_size}_${precision}_${run_process_type}_${run_mode}"
cmd="${python} ${BENCHMARK_ROOT}/scripts/analysis.py --filename ${log_path}/${log_name} \
--speed_log_file '${speed_log_path}/${speed_log_name}' \
--model_name ${_model_name} \
--base_batch_size ${batch_size} \
--run_mode ${run_mode} \
--run_process_type ${run_process_type} \
--fp_item ${precision} \
--keyword ips: \
--skip_steps 2 \
--device_num ${device_num} \
--speed_unit images/s \
--convergence_key loss: "
echo $cmd
eval $cmd
last_status=${PIPESTATUS[0]}
status_check $last_status "${cmd}" "${status_log}"
fi
done
done
done
\ No newline at end of file
tests/test_tipc/configs/conformer/train_benchmark.txt 0 → 100644
浏览文件 @ f3ec985a
===========================train_params===========================
model_name:conformer
python:python3.7
gpu_list:0|0,1
null:null
null:null
--benchmark-max-step:50
null:null
--benchmark-batch-size:16
null:null
null:null
null:null
null:null
##
trainer:norm_train
norm_train: ../paddlespeech/s2t/exps/u2/bin/train.py --config test_tipc/conformer/benchmark_train/conf/conformer.yaml --output test_tipc/conformer/benchmark_train/outputs --seed 1024
pact_train:null
fpgm_train:null
distill_train:null
null:null
null:null
##
===========================eval_params===========================
eval:null
null:null
##
===========================infer_params===========================
null:null
null:null
norm_export: null
quant_export:null
fpgm_export:null
distill_export:null
export1:null
export2:null
null:null
infer_model:null
infer_export:null
infer_quant:null
inference:null
null:null
null:null
null:null
null:null
null:null
null:null
null:null
null:null
null:null
null:null
null:null
===========================train_benchmark_params==========================
batch_size:16|30
fp_items:fp32
iteration:50
--profiler-options:"batch_range=[10,35];state=GPU;tracer_option=Default;profile_path=model.profile"
flags:FLAGS_eager_delete_tensor_gb=0.0;FLAGS_fraction_of_gpu_memory_to_use=0.98;FLAGS_conv_workspace_size_limit=4096"
tests/test_tipc/conformer/scripts/aishell_tiny.py 0 → 100644
浏览文件 @ f3ec985a
# Copyright (c) 2021 PaddlePaddle Authors. All Rights Reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
"""Prepare Aishell mandarin dataset
Download, unpack and create manifest files.
Manifest file is a json-format file with each line containing the
meta data (i.e. audio filepath, transcript and audio duration)
of each audio file in the data set.
"""
import argparse
import codecs
import json
import os
from pathlib import Path
import soundfile
from utils.utility import download
from utils.utility import unpack
DATA_HOME = os.path.expanduser('~/.cache/paddle/dataset/speech')
URL_ROOT_TAG
DATA_URL = URL_ROOT + '/data_aishell_tiny.tgz'
MD5_DATA = '337b1b1ea016761d4fd3225c5b8799b4'
RESOURCE_URL = URL_ROOT + '/resource_aishell.tgz'
MD5_RESOURCE = '957d480a0fcac85fc18e550756f624e5'
parser = argparse.ArgumentParser(description=__doc__)
parser.add_argument(
"--target_dir",
default=DATA_HOME + "/Aishell",
type=str,
help="Directory to save the dataset. (default: %(default)s)")
parser.add_argument(
"--manifest_prefix",
default="manifest",
type=str,
help="Filepath prefix for output manifests. (default: %(default)s)")
args = parser.parse_args()
def create_manifest(data_dir, manifest_path_prefix):
print("Creating manifest %s ..." % manifest_path_prefix)
json_lines = []
transcript_path = os.path.join(data_dir, 'transcript',
'aishell_transcript_v0.8.txt')
transcript_dict = {}
for line in codecs.open(transcript_path, 'r', 'utf-8'):
line = line.strip()
if line == '':
continue
audio_id, text = line.split(' ', 1)
# remove withespace, charactor text
text = ''.join(text.split())
transcript_dict[audio_id] = text
data_types = ['train', 'dev', 'test']
for dtype in data_types:
del json_lines[:]
total_sec = 0.0
total_text = 0.0
total_num = 0
audio_dir = os.path.join(data_dir, 'wav', dtype)
for subfolder, _, filelist in sorted(os.walk(audio_dir)):
for fname in filelist:
audio_path = os.path.abspath(os.path.join(subfolder, fname))
audio_id = os.path.basename(fname)[:-4]
# if no transcription for audio then skipped
if audio_id not in transcript_dict:
continue
utt2spk = Path(audio_path).parent.name
audio_data, samplerate = soundfile.read(audio_path)
duration = float(len(audio_data) / samplerate)
text = transcript_dict[audio_id]
json_lines.append(
json.dumps(
{
'utt': audio_id,
'utt2spk': str(utt2spk),
'feat': audio_path,
'feat_shape': (duration, ), # second
'text': text
},
ensure_ascii=False))
total_sec += duration
total_text += len(text)
total_num += 1
manifest_path = manifest_path_prefix + '.' + dtype
with codecs.open(manifest_path, 'w', 'utf-8') as fout:
for line in json_lines:
fout.write(line + '\n')
manifest_dir = os.path.dirname(manifest_path_prefix)
meta_path = os.path.join(manifest_dir, dtype) + '.meta'
with open(meta_path, 'w') as f:
print(f"{dtype}:", file=f)
print(f"{total_num} utts", file=f)
print(f"{total_sec / (60*60)} h", file=f)
print(f"{total_text} text", file=f)
print(f"{total_text / total_sec} text/sec", file=f)
print(f"{total_sec / total_num} sec/utt", file=f)
def prepare_dataset(url, md5sum, target_dir, manifest_path=None):
"""Download, unpack and create manifest file."""
data_dir = os.path.join(target_dir, 'data_aishell_tiny')
if not os.path.exists(data_dir):
filepath = download(url, md5sum, target_dir)
unpack(filepath, target_dir)
# unpack all audio tar files
audio_dir = os.path.join(data_dir, 'wav')
for subfolder, _, filelist in sorted(os.walk(audio_dir)):
for ftar in filelist:
unpack(os.path.join(subfolder, ftar), subfolder, True)
else:
print("Skip downloading and unpacking. Data already exists in %s." %
target_dir)
if manifest_path:
create_manifest(data_dir, manifest_path)
def main():
if args.target_dir.startswith('~'):
args.target_dir = os.path.expanduser(args.target_dir)
prepare_dataset(
url=DATA_URL,
md5sum=MD5_DATA,
target_dir=args.target_dir,
manifest_path=args.manifest_prefix)
prepare_dataset(
url=RESOURCE_URL,
md5sum=MD5_RESOURCE,
target_dir=args.target_dir,
manifest_path=None)
print("Data download and manifest prepare done!")
if __name__ == '__main__':
main()
tests/test_tipc/docs/benchmark_train.md 0 → 100644
浏览文件 @ f3ec985a
# TIPC Linux端Benchmark测试文档
该文档为Benchmark测试说明，Benchmark预测功能测试的主程序为`benchmark_train.sh`，用于验证监控模型训练的性能。
# 1. 测试流程
## 1.1 准备数据和环境安装
请在 repo根目录/tests 下运行
运行`test_tipc/prepare.sh`，完成训练数据准备和安装环境流程。
```shell
# 运行格式：bash test_tipc/prepare.sh train_benchmark.txt mode
bash test_tipc/prepare.sh test_tipc/configs/conformer/train_benchmark.txt benchmark_train
```
## 1.2 功能测试
执行`test_tipc/benchmark_train.sh`，完成模型训练和日志解析
```shell
# 运行格式：bash test_tipc/benchmark_train.sh train_benchmark.txt mode
bash test_tipc/benchmark_train.sh test_tipc/configs/conformer/train_benchmark.txt benchmark_train
```
`test_tipc/benchmark_train.sh`支持根据传入的第三个参数实现只运行某一个训练配置，如下：
```shell
# 运行格式：bash test_tipc/benchmark_train.sh train_benchmark.txt mode
bash test_tipc/benchmark_train.sh test_tipc/configs/conformer/train_benchmark.txt benchmark_train dynamic_bs16_fp32_DP_N1C1
```
dynamic_bs16_fp32_DP_N1C1为test_tipc/benchmark_train.sh传入的参数，格式如下：
`${modeltype}_${batch_size}_${fp_item}_${run_mode}_${device_num}`
包含的信息有：模型类型、batchsize大小、训练精度如fp32,fp16等、分布式运行模式以及分布式训练使用的机器信息如单机单卡（N1C1）。
## 2. 日志输出
运行后将保存模型的训练日志和解析日志，使用 `test_tipc/configs/conformer/train_benchmark.txt` 参数文件的训练日志解析结果是：
```
{"model_branch": "dygaph", "model_commit": "", "model_name": "conformer_bs16_fp32_SingleP_DP", "batch_size": 16, "fp_item": "fp32", "run_process_type": "SingleP", "run_mode": "DP", "convergence_value": "", "convergence_key": "loss:", "ips": , "speed_unit": "samples/s", "device_num": "N1C1", "model_run_time": "0", "frame_commit": "", "frame_version": "0.0.0"}
```
训练日志和日志解析结果保存在test目录下，文件组织格式如下：
```
test/
├── index
│   ├── tests_conformer_bs16_fp32_SingleP_DP_N1C1_speed
│   └── tests_conformer_bs16_fp32_SingleP_DP_N1C8_speed
├── profiling_log
│   └── tests_conformer_bs16_fp32_SingleP_DP_N1C1_profiling
└── train_log
├── tests_conformer_bs16_fp32_SingleP_DP_N1C1_log
└── tests_conformer_bs16_fp32_SingleP_DP_N1C8_log
```
tests/test_tipc/prepare.sh 0 → 100644
浏览文件 @ f3ec985a
#!/bin/bash
source test_tipc/common_func.sh
FILENAME=$1
# MODE be one of ['benchmark_train_lite_infer' 'benchmark_train_whole_infer' 'whole_train_whole_infer',
# 'whole_infer', 'klquant_whole_infer',
# 'cpp_infer', 'serving_infer', 'benchmark_train']
MODE=$2
dataline=$(cat ${FILENAME})
# parser params
IFS=$'\n'
lines=(${dataline})
# The training params
model_name=$(func_parser_value "${lines[1]}")
echo "model_name:"${model_name}
trainer_list=$(func_parser_value "${lines[14]}")
if [ ${MODE} = "benchmark_train" ];then
curPath=$(readlink -f "$(dirname "$0")")
echo "curPath:"${curPath}
cd ${curPath}/../..
pip install .
cd -
if [ ${model_name} == "conformer" ]; then
# set the URL for aishell_tiny dataset
URL='None'
echo "URL:"${URL}
if [ ${URL} == 'None' ];then
echo "please contact author to get the URL.\n"
exit
fi
sed -i "s#^URL_ROOT_TAG#URL_ROOT = '${URL}'#g" ${curPath}/conformer/scripts/aishell_tiny.py
cp ${curPath}/conformer/scripts/aishell_tiny.py ${curPath}/../../dataset/aishell/
cd ${curPath}/../../examples/aishell/asr1
source path.sh
# download audio data
sed -i "s#aishell.py#aishell_tiny.py#g" ./local/data.sh
bash ./local/data.sh || exit -1
if [ $? -ne 0 ]; then
exit 1
fi
mkdir -p ${curPath}/conformer/benchmark_train/
cp -rf conf ${curPath}/conformer/benchmark_train/
cp -rf data ${curPath}/conformer/benchmark_train/
cd ${curPath}
sed -i "s#accum_grad: 2#accum_grad: 1#g" ${curPath}/conformer/benchmark_train/conf/conformer.yaml
sed -i "s#data/#test_tipc/conformer/benchmark_train/data/#g" ${curPath}/conformer/benchmark_train/conf/conformer.yaml
sed -i "s#conf/#test_tipc/conformer/benchmark_train/conf/#g" ${curPath}/conformer/benchmark_train/conf/conformer.yaml
sed -i "s#data/#test_tipc/conformer/benchmark_train/data/#g" ${curPath}/conformer/benchmark_train/conf/tuning/decode.yaml
sed -i "s#data/#test_tipc/conformer/benchmark_train/data/#g" ${curPath}/conformer/benchmark_train/conf/preprocess.yaml
fi
fi
tests/test_tipc/test_train_inference_python.sh 0 → 100644
浏览文件 @ f3ec985a
#!/bin/bash
source test_tipc/common_func.sh
FILENAME=$1
# MODE be one of ['lite_train_lite_infer' 'lite_train_whole_infer' 'whole_train_whole_infer', 'whole_infer', 'klquant_whole_infer']
MODE=$2
dataline=$(awk 'NR==1, NR==51{print}' $FILENAME)
# parser params
IFS=$'\n'
lines=(${dataline})
# The training params
model_name=$(func_parser_value "${lines[1]}")
python=$(func_parser_value "${lines[2]}")
gpu_list=$(func_parser_value "${lines[3]}")
train_use_gpu_key=$(func_parser_key "${lines[4]}")
train_use_gpu_value=$(func_parser_value "${lines[4]}")
autocast_list=$(func_parser_value "${lines[5]}")
autocast_key=$(func_parser_key "${lines[5]}")
epoch_key=$(func_parser_key "${lines[6]}")
epoch_num=$(func_parser_params "${lines[6]}")
save_model_key=$(func_parser_key "${lines[7]}")
train_batch_key=$(func_parser_key "${lines[8]}")
train_batch_value=$(func_parser_params "${lines[8]}")
pretrain_model_key=$(func_parser_key "${lines[9]}")
pretrain_model_value=$(func_parser_value "${lines[9]}")
train_model_name=$(func_parser_value "${lines[10]}")
train_infer_img_dir=$(func_parser_value "${lines[11]}")
train_param_key1=$(func_parser_key "${lines[12]}")
train_param_value1=$(func_parser_value "${lines[12]}")
trainer_list=$(func_parser_value "${lines[14]}")
trainer_norm=$(func_parser_key "${lines[15]}")
norm_trainer=$(func_parser_value "${lines[15]}")
pact_key=$(func_parser_key "${lines[16]}")
pact_trainer=$(func_parser_value "${lines[16]}")
fpgm_key=$(func_parser_key "${lines[17]}")
fpgm_trainer=$(func_parser_value "${lines[17]}")
distill_key=$(func_parser_key "${lines[18]}")
distill_trainer=$(func_parser_value "${lines[18]}")
trainer_key1=$(func_parser_key "${lines[19]}")
trainer_value1=$(func_parser_value "${lines[19]}")
trainer_key2=$(func_parser_key "${lines[20]}")
trainer_value2=$(func_parser_value "${lines[20]}")
eval_py=$(func_parser_value "${lines[23]}")
eval_key1=$(func_parser_key "${lines[24]}")
eval_value1=$(func_parser_value "${lines[24]}")
save_infer_key=$(func_parser_key "${lines[27]}")
save_infer_value=$(func_parser_value "${lines[27]}")
export_weight=$(func_parser_key "${lines[28]}")
norm_export=$(func_parser_value "${lines[29]}")
pact_export=$(func_parser_value "${lines[30]}")
fpgm_export=$(func_parser_value "${lines[31]}")
distill_export=$(func_parser_value "${lines[32]}")
export_key1=$(func_parser_key "${lines[33]}")
export_value1=$(func_parser_value "${lines[33]}")
export_key2=$(func_parser_key "${lines[34]}")
export_value2=$(func_parser_value "${lines[34]}")
inference_dir=$(func_parser_value "${lines[35]}")
# parser inference model
infer_model_dir_list=$(func_parser_value "${lines[36]}")
infer_export_list=$(func_parser_value "${lines[37]}")
infer_is_quant=$(func_parser_value "${lines[38]}")
# parser inference
inference_py=$(func_parser_value "${lines[39]}")
use_gpu_key=$(func_parser_key "${lines[40]}")
use_gpu_list=$(func_parser_value "${lines[40]}")
use_mkldnn_key=$(func_parser_key "${lines[41]}")
use_mkldnn_list=$(func_parser_value "${lines[41]}")
cpu_threads_key=$(func_parser_key "${lines[42]}")
cpu_threads_list=$(func_parser_value "${lines[42]}")
batch_size_key=$(func_parser_key "${lines[43]}")
batch_size_list=$(func_parser_value "${lines[43]}")
use_trt_key=$(func_parser_key "${lines[44]}")
use_trt_list=$(func_parser_value "${lines[44]}")
precision_key=$(func_parser_key "${lines[45]}")
precision_list=$(func_parser_value "${lines[45]}")
infer_model_key=$(func_parser_key "${lines[46]}")
image_dir_key=$(func_parser_key "${lines[47]}")
infer_img_dir=$(func_parser_value "${lines[47]}")
save_log_key=$(func_parser_key "${lines[48]}")
benchmark_key=$(func_parser_key "${lines[49]}")
benchmark_value=$(func_parser_value "${lines[49]}")
infer_key1=$(func_parser_key "${lines[50]}")
infer_value1=$(func_parser_value "${lines[50]}")
# parser klquant_infer
if [ ${MODE} = "klquant_whole_infer" ]; then
dataline=$(awk 'NR==1 NR==17{print}' $FILENAME)
lines=(${dataline})
model_name=$(func_parser_value "${lines[1]}")
python=$(func_parser_value "${lines[2]}")
# parser inference model
infer_model_dir_list=$(func_parser_value "${lines[3]}")
infer_export_list=$(func_parser_value "${lines[4]}")
infer_is_quant=$(func_parser_value "${lines[5]}")
# parser inference
inference_py=$(func_parser_value "${lines[6]}")
use_gpu_key=$(func_parser_key "${lines[7]}")
use_gpu_list=$(func_parser_value "${lines[7]}")
use_mkldnn_key=$(func_parser_key "${lines[8]}")
use_mkldnn_list=$(func_parser_value "${lines[8]}")
cpu_threads_key=$(func_parser_key "${lines[9]}")
cpu_threads_list=$(func_parser_value "${lines[9]}")
batch_size_key=$(func_parser_key "${lines[10]}")
batch_size_list=$(func_parser_value "${lines[10]}")
use_trt_key=$(func_parser_key "${lines[11]}")
use_trt_list=$(func_parser_value "${lines[11]}")
precision_key=$(func_parser_key "${lines[12]}")
precision_list=$(func_parser_value "${lines[12]}")
infer_model_key=$(func_parser_key "${lines[13]}")
image_dir_key=$(func_parser_key "${lines[14]}")
infer_img_dir=$(func_parser_value "${lines[14]}")
save_log_key=$(func_parser_key "${lines[15]}")
benchmark_key=$(func_parser_key "${lines[16]}")
benchmark_value=$(func_parser_value "${lines[16]}")
infer_key1=$(func_parser_key "${lines[17]}")
infer_value1=$(func_parser_value "${lines[17]}")
fi
save_model_value=$(func_parser_value "${lines[7]}")
if [[ ${save_model_value} = " " ]] || [[ ${save_model_value} = "null" ]] || [[ ${save_model_value} = "" ]];then
LOG_PATH="./test_tipc/output"
else
LOG_PATH=${save_model_value}
fi
mkdir -p ${LOG_PATH}
status_log="${LOG_PATH}/results_python.log"
function func_inference(){
IFS='|'
_python=$1
_script=$2
_model_dir=$3
_log_path=$4
_img_dir=$5
_flag_quant=$6
# inference
for use_gpu in ${use_gpu_list[*]}; do
if [ ${use_gpu} = "False" ] || [ ${use_gpu} = "cpu" ]; then
for use_mkldnn in ${use_mkldnn_list[*]}; do
if [ ${use_mkldnn} = "False" ] && [ ${_flag_quant} = "True" ]; then
continue
fi
for threads in ${cpu_threads_list[*]}; do
for batch_size in ${batch_size_list[*]}; do
for precision in ${precision_list[*]}; do
if [ ${use_mkldnn} = "False" ] && [ ${precision} = "fp16" ]; then
continue
fi # skip when enable fp16 but disable mkldnn
if [ ${_flag_quant} = "True" ] && [ ${precision} != "int8" ]; then
continue
fi # skip when quant model inference but precision is not int8
set_precision=$(func_set_params "${precision_key}" "${precision}")
_save_log_path="${_log_path}/python_infer_cpu_usemkldnn_${use_mkldnn}_threads_${threads}_precision_${precision}_batchsize_${batch_size}.log"
set_infer_data=$(func_set_params "${image_dir_key}" "${_img_dir}")
set_benchmark=$(func_set_params "${benchmark_key}" "${benchmark_value}")
set_batchsize=$(func_set_params "${batch_size_key}" "${batch_size}")
set_cpu_threads=$(func_set_params "${cpu_threads_key}" "${threads}")
set_model_dir=$(func_set_params "${infer_model_key}" "${_model_dir}")
set_infer_params1=$(func_set_params "${infer_key1}" "${infer_value1}")
command="${_python} ${_script} ${use_gpu_key}=${use_gpu} ${use_mkldnn_key}=${use_mkldnn} ${set_cpu_threads} ${set_model_dir} ${set_batchsize} ${set_infer_data} ${set_benchmark} ${set_precision} ${set_infer_params1} > ${_save_log_path} 2>&1 "
eval $command
last_status=${PIPESTATUS[0]}
eval "cat ${_save_log_path}"
status_check $last_status "${command}" "${status_log}"
done
done
done
done
elif [ ${use_gpu} = "True" ] || [ ${use_gpu} = "gpu" ]; then
for use_trt in ${use_trt_list[*]}; do
for precision in ${precision_list[*]}; do
if [[ ${_flag_quant} = "False" ]] && [[ ${precision} =~ "int8" ]]; then
continue
fi
if [[ ${precision} =~ "fp16" || ${precision} =~ "int8" ]] && [ ${use_trt} = "False" ]; then
continue
fi
if [[ ${use_trt} = "False" || ${precision} =~ "int8" ]] && [ ${_flag_quant} = "True" ]; then
continue
fi
for batch_size in ${batch_size_list[*]}; do
_save_log_path="${_log_path}/python_infer_gpu_usetrt_${use_trt}_precision_${precision}_batchsize_${batch_size}.log"
set_infer_data=$(func_set_params "${image_dir_key}" "${_img_dir}")
set_benchmark=$(func_set_params "${benchmark_key}" "${benchmark_value}")
set_batchsize=$(func_set_params "${batch_size_key}" "${batch_size}")
set_tensorrt=$(func_set_params "${use_trt_key}" "${use_trt}")
set_precision=$(func_set_params "${precision_key}" "${precision}")
set_model_dir=$(func_set_params "${infer_model_key}" "${_model_dir}")
set_infer_params1=$(func_set_params "${infer_key1}" "${infer_value1}")
command="${_python} ${_script} ${use_gpu_key}=${use_gpu} ${set_tensorrt} ${set_precision} ${set_model_dir} ${set_batchsize} ${set_infer_data} ${set_benchmark} ${set_infer_params1} > ${_save_log_path} 2>&1 "
eval $command
last_status=${PIPESTATUS[0]}
eval "cat ${_save_log_path}"
status_check $last_status "${command}" "${status_log}"
done
done
done
else
echo "Does not support hardware other than CPU and GPU Currently!"
fi
done
}
if [ ${MODE} = "whole_infer" ] || [ ${MODE} = "klquant_whole_infer" ]; then
GPUID=$3
if [ ${#GPUID} -le 0 ];then
env=" "
else
env="export CUDA_VISIBLE_DEVICES=${GPUID}"
fi
# set CUDA_VISIBLE_DEVICES
eval $env
export Count=0
IFS="|"
infer_run_exports=(${infer_export_list})
infer_quant_flag=(${infer_is_quant})
for infer_model in ${infer_model_dir_list[*]}; do
# run export
if [ ${infer_run_exports[Count]} != "null" ];then
save_infer_dir=$(dirname $infer_model)
set_export_weight=$(func_set_params "${export_weight}" "${infer_model}")
set_save_infer_key=$(func_set_params "${save_infer_key}" "${save_infer_dir}")
# For lac which needs the `data_dir` args
set_export1_key=$(func_set_params "${export_key1}" "${export_value1}")
export_cmd="${python} ${infer_run_exports[Count]} ${set_export_weight} ${set_save_infer_key} ${set_export1_key}"
echo ${infer_run_exports[Count]}
echo $export_cmd
eval $export_cmd
status_export=$?
status_check $status_export "${export_cmd}" "${status_log}"
else
save_infer_dir=${infer_model}
fi
#run inference
is_quant=${infer_quant_flag[Count]}
if [ ${MODE} = "klquant_infer" ]; then
is_quant="True"
fi
func_inference "${python}" "${inference_py}" "${save_infer_dir}" "${LOG_PATH}" "${infer_img_dir}" ${is_quant}
Count=$(($Count + 1))
done
else
IFS="|"
export Count=0
USE_GPU_KEY=(${train_use_gpu_value})
for gpu in ${gpu_list[*]}; do
train_use_gpu=${USE_GPU_KEY[Count]}
Count=$(($Count + 1))
ips=""
if [ ${gpu} = "-1" ];then
env=""
elif [ ${#gpu} -le 1 ];then
env="export CUDA_VISIBLE_DEVICES=${gpu}"
eval ${env}
elif [ ${#gpu} -le 15 ];then
IFS=","
array=(${gpu})
env="export CUDA_VISIBLE_DEVICES=${array[0]}"
IFS="|"
else
IFS=";"
array=(${gpu})
ips=${array[0]}
gpu=${array[1]}
IFS="|"
env=" "
fi
for autocast in ${autocast_list[*]}; do
if [ ${autocast} = "amp" ]; then
set_amp_config="Global.use_amp=True Global.scale_loss=1024.0 Global.use_dynamic_loss_scaling=True"
else
set_amp_config=" "
fi
for trainer in ${trainer_list[*]}; do
flag_quant=False
if [ ${trainer} = ${pact_key} ]; then
run_train=${pact_trainer}
run_export=${pact_export}
flag_quant=True
elif [ ${trainer} = "${fpgm_key}" ]; then
run_train=${fpgm_trainer}
run_export=${fpgm_export}
elif [ ${trainer} = "${distill_key}" ]; then
run_train=${distill_trainer}
run_export=${distill_export}
elif [ ${trainer} = ${trainer_key1} ]; then
run_train=${trainer_value1}
run_export=${export_value1}
elif [[ ${trainer} = ${trainer_key2} ]]; then
run_train=${trainer_value2}
run_export=${export_value2}
else
run_train=${norm_trainer}
run_export=${norm_export}
fi
if [ ${run_train} = "null" ]; then
continue
fi
set_autocast=$(func_set_params "${autocast_key}" "${autocast}")
set_epoch=$(func_set_params "${epoch_key}" "${epoch_num}")
set_pretrain=$(func_set_params "${pretrain_model_key}" "${pretrain_model_value}")
set_batchsize=$(func_set_params "${train_batch_key}" "${train_batch_value}")
set_train_params1=$(func_set_params "${train_param_key1}" "${train_param_value1}")
set_use_gpu=$(func_set_params "${train_use_gpu_key}" "${train_use_gpu}")
if [ ${#ips} -le 26 ];then
save_log="${LOG_PATH}/${trainer}_gpus_${gpu}_autocast_${autocast}"
nodes=1
else
IFS=","
ips_array=(${ips})
IFS="|"
nodes=${#ips_array[@]}
save_log="${LOG_PATH}/${trainer}_gpus_${gpu}_autocast_${autocast}_nodes_${nodes}"
fi
# load pretrain from norm training if current trainer is pact or fpgm trainer
if ([ ${trainer} = ${pact_key} ] || [ ${trainer} = ${fpgm_key} ]) && [ ${nodes} -le 1 ]; then
set_pretrain="${load_norm_train_model}"
fi
set_save_model=$(func_set_params "${save_model_key}" "${save_log}")
if [ ${#gpu} -le 2 ];then # train with cpu or single gpu
cmd="${python} ${run_train} ${set_use_gpu} ${set_save_model} ${set_epoch} ${set_pretrain} ${set_autocast} ${set_batchsize} ${set_train_params1} ${set_amp_config} "
elif [ ${#ips} -le 26 ];then # train with multi-gpu
cmd="${python} -m paddle.distributed.launch --gpus=${gpu} ${run_train} ${set_use_gpu} ${set_save_model} ${set_epoch} ${set_pretrain} ${set_autocast} ${set_batchsize} ${set_train_params1} ${set_amp_config}"
else # train with multi-machine
cmd="${python} -m paddle.distributed.launch --ips=${ips} --gpus=${gpu} ${run_train} ${set_use_gpu} ${set_save_model} ${set_pretrain} ${set_epoch} ${set_autocast} ${set_batchsize} ${set_train_params1} ${set_amp_config}"
fi
# run train
eval "unset CUDA_VISIBLE_DEVICES"
eval $cmd
status_check $? "${cmd}" "${status_log}"
set_eval_pretrain=$(func_set_params "${pretrain_model_key}" "${save_log}/${train_model_name}")
# save norm trained models to set pretrain for pact training and fpgm training
if [ ${trainer} = ${trainer_norm} ] && [ ${nodes} -le 1 ]; then
load_norm_train_model=${set_eval_pretrain}
fi
# run eval
if [ ${eval_py} != "null" ]; then
set_eval_params1=$(func_set_params "${eval_key1}" "${eval_value1}")
eval_cmd="${python} ${eval_py} ${set_eval_pretrain} ${set_use_gpu} ${set_eval_params1}"
eval $eval_cmd
status_check $? "${eval_cmd}" "${status_log}"
fi
# run export model
if [ ${run_export} != "null" ]; then
# run export model
save_infer_path="${save_infer_value}"
set_export_weight=$(func_set_params "${export_weight}" "${save_log}/${train_model_name}")
set_save_infer_key=$(func_set_params "${save_infer_key}" "${save_infer_path}")
# For lac which needs the `data_dir` args
set_export1_key=$(func_set_params "${export_key1}" "${export_value1}")
export_cmd="${python} ${run_export} ${set_export_weight} ${set_save_infer_key} ${set_export1_key}"
eval $export_cmd
status_check $? "${export_cmd}" "${status_log}"
#run inference
eval $env
save_infer_path="${save_infer_value}"
if [ ${inference_dir} != "null" ] && [ ${inference_dir} != '##' ]; then
infer_model_dir="${save_infer_path}/${inference_dir}"
else
infer_model_dir=${save_infer_path}
fi
func_inference "${python}" "${inference_py}" "${infer_model_dir}" "${LOG_PATH}" "${train_infer_img_dir}" "${flag_quant}"
eval "unset CUDA_VISIBLE_DEVICES"
fi
done # done with: for trainer in ${trainer_list[*]}; do
done # done with: for autocast in ${autocast_list[*]}; do
done # done with: for gpu in ${gpu_list[*]}; do
fi # end if [ ${MODE} = "infer" ]; then
Markdown is supported
0% 或 .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑！
想要评论请 注册 或